|

Highlights & scope

• >Lead development of an autonomous code review agent using LangChain Agents and OpenAI GPT-4o on FastAPI, orchestrating parallel security and performance scans that reduce PR turnaround from 18 h to 2.5 h.
• >Engineer event-driven webhook pipeline in Node.js and Express within a microservices architecture, processing GitHub PR events and dispatching payloads to the Python agent service via REST APIs, enabling real-time reviews for 200 engineers.
• >Implement RAG-based codebase search using a vector database and LangChain retrieval chains integrated with the GitHub REST API, providing contextual diff analysis that identifies 34 critical security vulnerabilities missed during manual review cycles.
• >Develop confidence-scored suggestion engine with automated test generation using LangChain structured outputs, mentoring 3 juniors on evaluation patterns while persisting audit trails in PostgreSQL with SQL analytics achieving 67% acceptance rate.
• >Build admin analytics dashboard in React and Next.js with product and engineering leadership to define acceptance-rate, false-positive, and cost-per-PR metrics, reducing review cost from $50 to $0.40 per PR.
• >Containerize microservices with Docker and deploy to AWS ECS, Lambda, and Kubernetes (EKS), integrating Slack REST API notifications with risk scoring that improves developer satisfaction from 3.2 to 4.6 out of 5.0.
• >Orchestrate end-to-end CI/CD pipelines using GitHub Actions with automated integration tests and agent quality gates, deploying across AWS infrastructure in Agile sprints with 99.8% pipeline reliability processing 600+ monthly PRs.

Highlights & scope

• >Led a 4-engineer squad to architect payment processing core in Java, Spring Boot, and Hibernate, implementing intelligent retry and multi-gateway routing across UPI rails reducing failure rates from 12% to 4%, recovering $2.1M annually.
• >Drove system design of event-driven microservices using Apache Kafka, streaming payment events, settlement triggers, and refund workflows across 6 consumer services processing 500K+ daily transactions with guaranteed ordering.
• >Built settlement reconciliation engine in Python and FastAPI, consuming Kafka events and automating bank file matching against PostgreSQL and MySQL ledgers with optimized indexing, reducing reconciliation turnaround from 48 h to 2 h.
• >Developed merchant transaction dashboard in React and TypeScript with real-time payment tracking, settlement summaries, and GMV analytics via REST APIs, collaborating with product and design to serve 25K+ merchants.
• >Engineered refund orchestration and webhook services in Node.js and Express, managing state machines with idempotency keys and Redis-backed rate limiting via REST APIs, reducing refund processing from 7 days to under 24 h.
• >Established comprehensive JUnit test suites achieving 90%+ coverage across Spring Boot payment-critical services, containerized with Docker, and deployed to AWS ECS via Jenkins and GitHub Actions CI/CD pipelines in Agile sprints.
• >Streamed transaction lifecycle events from Kafka into DynamoDB audit trails with TTL retention, enforced Git review gates with mandatory approvals for payment-critical paths, enabling PCI-DSS compliance across 500K+ daily transactions.